A Comparative Study of Feature Fusion Strategies for Hyperspectral and LiDAR Remote Sensing Data
Keywords:
hyperspectral, LiDAR, feature fusion, remote sensing, multi-modal data, system architecture, governanceAbstract
The integration of hyperspectral imaging and Light Detection and Ranging (LiDAR) data has emerged as a powerful paradigm for land cover classification, environmental monitoring, and urban infrastructure assessment. While each modality offers complementary strengths—hyperspectral sensors capture detailed spectral signatures across hundreds of narrow bands, and LiDAR provides precise three-dimensional structural information—their fusion presents significant architectural and algorithmic challenges. This paper presents a comparative study of feature fusion strategies employed in the joint processing of hyperspectral and LiDAR remote sensing data, focusing on early, intermediate, and late fusion paradigms. Rather than emphasizing algorithmic novelty, the analysis centers on systemic trade-offs related to computational efficiency, spatial and spectral alignment, scalability, and interpretability. The study examines how different fusion architectures affect system robustness, fairness in classification across diverse land cover classes, and deployment feasibility in operational remote sensing infrastructures. Particular attention is given to the role of band ordering and its impact on feature representation, a topic explored in recent work that evaluates ordering strategies in hyperspectral-LiDAR fusion networks. Through a synthesis of contemporary literature and conceptual analysis, this paper outlines governance implications for large-scale geospatial data systems, including data standardization, model transparency, and policy-driven calibration requirements. The findings suggest that no single fusion strategy universally outperforms others; instead, optimal design depends on the specific application domain, data quality constraints, and institutional priorities regarding interpretability versus predictive accuracy. The paper concludes with forward-looking perspectives on sustainable fusion frameworks that balance performance with accountability in increasingly automated remote sensing pipelines.
References
1. Ghamisi, P., Yokoya, N., Li, J., Liao, W., Liu, S., Plaza, J., Rasti, B., & Plaza, A. (2017). Advances in hyperspectral image and signal processing: A comprehensive overview of the state of the art. IEEE Geoscience and Remote Sensing Magazine, 5(4), 37–78.
2. Yokoya, N., Ghamisi, P., Xia, J., Sukhanov, S., & Herold, M. (2017). Open data for global multimodal land use classification: A review. IEEE Geoscience and Remote Sensing Magazine, 5(4), 79–99.
3. Rasti, B., Ghamisi, P., & Gloaguen, R. (2020). Hyperspectral and LiDAR fusion using deep learning: A review. Remote Sensing, 12(6), 987.
4. Hong, D., Gao, L., Yokoya, N., Yao, J., Chanussot, J., Du, Q., & Zhang, B. (2019). More diverse means better: Multimodal deep learning meets remote sensing imagery classification. IEEE Transactions on Geoscience and Remote Sensing, 59(5), 4340–4354.
5. Bioucas-Dias, J. M., Plaza, A., Camps-Valls, G., Scheunders, P., Nasrabadi, N. M., & Chanussot, J. (2013). Hyperspectral remote sensing data analysis and future challenges. IEEE Geoscience and Remote Sensing Magazine, 1(2), 6–36.
6. Dubayah, R., & Drake, J. B. (2000). The use of lidar remote sensing to estimate biomass in deciduous forests. Journal of Forestry, 98(3), 44–48.
7. Dalponte, M., Bruzzone, L., & Gianelle, D. (2008). Fusion of hyperspectral and LIDAR remote sensing data for classification of complex forest areas. IEEE Transactions on Geoscience and Remote Sensing, 46(5), 1416–1427.
8. Li, Y., Zhang, H., & Shen, Q. (2017). Spectral-spatial classification of hyperspectral imagery with 3D convolutional neural network. Remote Sensing, 9(1), 67.
9. Ghamisi, P., Höfle, B., & Zhu, X. X. (2019). Hyperspectral and LiDAR fusion with deep learning: A new paradigm for land cover classification. IEEE Geoscience and Remote Sensing Letters, 16(7), 1056–1060.
10. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 770–778.
11. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30, 5998–6008.
12. Zhang, L., Zhang, L., & Du, B. (2016). Deep learning for remote sensing data: A technical tutorial on the state of the art. IEEE Geoscience and Remote Sensing Magazine, 4(2), 22–40.
13. Sun, X., & Fang, L. (2020). Hyperspectral and LiDAR data fusion: A multiscale framework using sparse representation and guided filtering. IEEE Transactions on Geoscience and Remote Sensing, 58(5), 3402–3415.
14. Montavon, G., Samek, W., & Müller, K.-R. (2018). Methods for interpreting and understanding deep neural networks. Digital Signal Processing, 73, 1–15.
15. Yang, J. X., Wang, J., Li, Z., Sui, C., Long, Z., & Zhou, J. (2025). HSLiNets: Evaluating Band Ordering Strategies in Hyperspectral and LiDAR Fusion. IEEE Geoscience and Remote Sensing Letters.
16. Xu, X., Li, W., Ran, Q., Du, Q., & Gao, L. (2019). Multisource remote sensing data classification based on convolutional neural network. IEEE Transactions on Geoscience and Remote Sensing, 57(12), 10354–10365.
17. Zhang, Y., & Ma, J. (2021). A cross-modal attention-guided fusion network for hyperspectral and LiDAR data classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 14, 8200–8212.
18. Ghamisi, P., & Yokoya, N. (2018). IMG2DSM: Height estimation from single remote sensing images using deep learning. IEEE Geoscience and Remote Sensing Letters, 15(11), 1705–1709.
19. Huang, Y., & Chen, Z. (2022). Robust fusion of hyperspectral and LiDAR data under misalignment: A feature pyramid approach. ISPRS Journal of Photogrammetry and Remote Sensing, 186, 213–226.
20. Arvor, D., Dube, F., & Lelong, C. (2021). Geospatial data governance: Challenges and opportunities for earth observation. Big Earth Data, 5(2), 189–207.
21. Zhu, X. X., Tuia, D., Mou, L., Xia, G.-S., Zhang, L., Xu, F., & Fraundorfer, F. (2017). Deep learning in remote sensing: A comprehensive review and list of resources. IEEE Geoscience and Remote Sensing Magazine, 5(4), 8–36.
22. Barocas, S., Hardt, M., & Narayanan, A. (2019). Fairness and Machine Learning. MIT Press.
23. Rolnick, D., Donti, P. L., Kaack, L. H., Kochanski, K., Lacoste, A., Sankaran, K., Ross, A. S., Milojevic-Dupont, N., Jaques, N., Waldman-Brown, A., Luccioni, A. S., Maharaj, T., Sherwin, E. D., Mukkavilli, S. K., Kording, K. P., Gomes, C. P., Ng, A. Y., Hassabis, D., & Bengio, Y. (2022). Tackling climate change with machine learning. ACM Computing Surveys, 55(2), 1–96.
24. Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., & Houlsby, N. (2021). An image is worth 16x16 words: Transformers for image recognition at scale. International Conference on Learning Representations.
25. Doshi-Velez, F., & Kim, B. (2017). Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 International Journal of Artificial Intelligence Research
This work is licensed under a Creative Commons Attribution 4.0 International License.
This article is published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
